{"id":243,"date":"2017-03-25T16:00:58","date_gmt":"2017-03-25T16:00:58","guid":{"rendered":"https:\/\/marriott-stats.com\/nigels-blog\/?p=243"},"modified":"2019-03-28T09:15:59","modified_gmt":"2019-03-28T09:15:59","slug":"stats-in-the-news-1-science-technology-march-2017","status":"publish","type":"post","link":"https:\/\/marriott-stats.com\/nigels-blog\/stats-in-the-news-1-science-technology-march-2017\/","title":{"rendered":"Stats in the News #1 &#8211; Science &amp; Technology March 2017"},"content":{"rendered":"<p>Welcome to my first post where I put my <a href=\"https:\/\/marriott-stats.com\/nigels-blog\/stats-in-the-news-0-the-evidence-hierarchy-and-how-to-use-it\/\" target=\"_blank\" rel=\"noopener noreferrer\">Evidence Hierarchy or Circle\u00a0<\/a>into practice and show you what is behind the headline.<img loading=\"lazy\" decoding=\"async\" class=\"alignright size-medium wp-image-216\" src=\"https:\/\/marriott-stats.com\/nigels-blog\/wp-content\/uploads\/2017\/03\/Evidence2-300x190.png\" alt=\"\" width=\"300\" height=\"190\" srcset=\"https:\/\/marriott-stats.com\/nigels-blog\/wp-content\/uploads\/2017\/03\/Evidence2-300x190.png 300w, https:\/\/marriott-stats.com\/nigels-blog\/wp-content\/uploads\/2017\/03\/Evidence2-768x487.png 768w, https:\/\/marriott-stats.com\/nigels-blog\/wp-content\/uploads\/2017\/03\/Evidence2-450x285.png 450w, https:\/\/marriott-stats.com\/nigels-blog\/wp-content\/uploads\/2017\/03\/Evidence2.png 1013w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/p>\n<p>Today I am concentrating on science and technology related articles from the BBC website since that is accessible to nearly everyone.\u00a0 As always, I am critiquing the article more than the research since I have not read the research papers that motivated the article.\u00a0 The 3 articles are:<\/p>\n<ol>\n<li>&#8220;<a href=\"http:\/\/www.bbc.co.uk\/news\/technology-39364974\" target=\"_blank\" rel=\"noopener noreferrer\">Fruit shaped sensor can improve freshness<\/a>&#8220;.<\/li>\n<li>&#8220;<a href=\"http:\/\/www.bbc.co.uk\/news\/business-39377353\" target=\"_blank\" rel=\"noopener noreferrer\">Robots to affect up to 30% of jobs<\/a>&#8220;<\/li>\n<li>&#8220;<a href=\"http:\/\/www.bbc.co.uk\/news\/science-environment-39305750\" target=\"_blank\" rel=\"noopener noreferrer\">Dinosaurs may have UK origin<\/a>&#8220;<\/li>\n<\/ol>\n<h3><!--more--><strong><a href=\"http:\/\/www.bbc.co.uk\/news\/technology-39364974\" target=\"_blank\" rel=\"noopener noreferrer\">&#8220;Fruit shaped sensor can improve freshness<\/a>&#8220;<\/strong><\/h3>\n<p>This is a good <em><strong>Belief&gt;Logic&gt;Experiment<\/strong><\/em> example to start with.\u00a0 The article\u00a0describes the development of an artificial fruit with sensors that could be loaded alongside real fruit when they are being shipped around the world.\u00a0 The sensor would monitor temperatures as it is shipped and\u00a0the shippers would be able to take action\u00a0if temperatures were reaching levels that might spoil the fruit.\u00a0 The goal is to reduce spoilage of fruit in shipments.<\/p>\n<p>If I apply the Evidence Circle to this article, this is what I see:<\/p>\n<ol>\n<li>No data is provided in the article\u00a0on how much fruit is spoiled today after shipment.\u00a0 Such a figure would give us an idea of the scale of the problem and the potential benefit of a solution to the problem.\u00a0 Without that information, we are left with a BELIEF that fruit spoilage is a problem that needs a solution.\u00a0 I am sure that it is a major problem and that data is available to measure the scale of the problem but the article does not give that information.\u00a0 We also told that spoilage is linked with temperature but again no information is given on how this relationship works i.e. what is a good range for fruit storage.<\/li>\n<li>From this belief that fruit spoilage is a problem and that it is correlated with temperature, LOGIC has been used to determine that if we could measure the temperature within a shipment of fruit (rather than the air temperature surrounding the fruit at present), we would be able to control shipment temperatures a lot better and this would result in lower spoilage.\u00a0 The end result is the proposed sensor intended to mimic certain fruit items.<\/li>\n<li>Currently a prototype is being built and the intention is to test it.\u00a0 This will require an EXPERIMENT to see if the sensor is capable of measuring the right temperature and transmitting that information to a shipper.\u00a0 The experiment should be designed so that it can measure the benefit of the sensor i.e. by how much could fruit spoilage be reduced.<\/li>\n<\/ol>\n<p>Together I see within the article a chain from BELIEF to EXPERIMENT which is something that does occur in a lot of R&amp;D departments i.e. identify a problem, figure out a possible solution and test that the solution works.\u00a0 Actually it is possible to anticipate that if the experiment is successful, the next stage will Observation and Anecdote as follows:<\/p>\n<ul>\n<li>A successful experiment should see this device rolled out.\u00a0 Shippers will then be able to OBSERVE if their fruit spoilage rates have fallen and are reaching the levels predicted by the experiment.<\/li>\n<li>If lower fruit spoilage levels are achieved, then supermarkets would hope hear their customers say how good they are at delivering fresh fruit i.e. ANECDOTES.<\/li>\n<\/ul>\n<h3><strong>&#8220;<a href=\"http:\/\/www.bbc.co.uk\/news\/business-39377353\" target=\"_blank\" rel=\"noopener noreferrer\">Robots to affect up to 30% of jobs<\/a>&#8220;<\/strong><\/h3>\n<p>This is a study that attempts to estimate what % of jobs could be automated in the future.\u00a0 At the end, a list of sectors is given along with the estimated % at risk of automation.\u00a0 This list varies from 56% in the Transport sector to 9% in the Education sector.\u00a0 Since I am a <a href=\"https:\/\/marriott-stats.com\/training\/\" target=\"_blank\" rel=\"noopener noreferrer\">statistical trainer <\/a>rather than a train driver, it seems that the risk of me being automated out of a job is somewhat low but how much weight should I give to this article?<\/p>\n<p>The article is a combination of BELIEFS and OBSERVATIONS.\u00a0 Although not explicitly stated in the article, I would assume that PwC took employment data that is normally available from the Office of National Statistics that breaks down the number of jobs by type of industry and occupation.\u00a0 Such data would be considered observational and is tracked over time by the ONS.\u00a0 The next step would have been to estimated the likelihood of each type of job being automated in the near future.\u00a0 Such estimates would have to be based on beliefs even if the belief comes from an expert in those industries.\u00a0 Some anecdotal evidence was probably used as well to refine the probability of a specific type of job being automated.\u00a0 Combine all this and you end up with the table shown.<\/p>\n<p>In a later post, I will be talking about <a href=\"https:\/\/marriott-stats.com\/nigels-blog\/forecasting1-how-do-you-identify-a-good-forecaster\/\" target=\"_blank\" rel=\"noopener noreferrer\">what makes a good forecaster <\/a>but projecting the future is always fraught with danger.\u00a0 This type of article is quite a common one in my experience where the research has made use of Observational evidence but along the way, a belief or series or beliefs has been slipped in somewhere and the whole premise of the article depends on that belief being correct.<\/p>\n<h3><strong>&#8220;<a href=\"http:\/\/www.bbc.co.uk\/news\/science-environment-39305750\" target=\"_blank\" rel=\"noopener noreferrer\">Dinosaurs may have UK origin<\/a>&#8220;<\/strong><\/h3>\n<p>The entire study of Dinosaurs is essentially one of ANECDOTES.\u00a0 The basic problem is that we can only study the fossils that exist today and\u00a0have been dug up.\u00a0 Only a tiny fraction of all the dinosaurs that ever lived will end up being studied as a fossil and we have no control over which dinosaur becomes a fossil.\u00a0 We have some control over where to go looking for fossils but the end result is that we end up with a very patchy, fragmented and biased sample of data.\u00a0 So it is not surprising that as new anecdotal evidence becomes available, the consensus might change.<\/p>\n<p>This article describes such a change of consensus with the meat eaters now being classified as bird-hipped rather than lizard-hipped.\u00a0 Classification of fossils is a major science and involves characterising and then grouping fossils.\u00a0 This approach uses a lot of statistics that come under the banner of multivariate analysis and is one of my favourite areas of statistics.\u00a0 Instead of fossils, most of my work involves living human beings being classified under the banner of <a href=\"https:\/\/marriott-stats.com\/surveys\/\" target=\"_blank\" rel=\"noopener noreferrer\">customer segmentation<\/a>.<\/p>\n<p>The upshot of the reclassification is that two fossils found in Britain are now thought to be some of the oldest dinosaur fossils leading to the speculation that dinosaurs originated in the UK.\u00a0 A sample size of 2 collected under highly uncontrolled conditions barely qualifies as anecdotal evidence and I am not surprised that an American scientist quoted in the article comes across as sceptical.\u00a0 Will the origins of dinosaurs become the new Anglo-American rivalry?<\/p>\n<p>&nbsp;<\/p>\n<p>&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Welcome to my first post where I put my Evidence Hierarchy or Circle\u00a0into practice and show you what is behind the headline. Today I am concentrating on science and technology related articles from the BBC website since that is accessible to nearly everyone.\u00a0 As always, I am critiquing the article more than the research since [&hellip;]<\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_genesis_hide_title":false,"_genesis_hide_breadcrumbs":false,"_genesis_hide_singular_image":false,"_genesis_hide_footer_widgets":false,"_genesis_custom_body_class":"","_genesis_custom_post_class":"","_genesis_layout":"","footnotes":""},"categories":[1],"tags":[37,36,35,38],"class_list":{"0":"post-243","1":"post","2":"type-post","3":"status-publish","4":"format-standard","6":"category-misc","7":"tag-data-journalism","8":"tag-evidence","9":"tag-evidence-hierarchy","10":"tag-stats-in-the-news","11":"entry","12":"override"},"_links":{"self":[{"href":"https:\/\/marriott-stats.com\/nigels-blog\/wp-json\/wp\/v2\/posts\/243","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/marriott-stats.com\/nigels-blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/marriott-stats.com\/nigels-blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/marriott-stats.com\/nigels-blog\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/marriott-stats.com\/nigels-blog\/wp-json\/wp\/v2\/comments?post=243"}],"version-history":[{"count":5,"href":"https:\/\/marriott-stats.com\/nigels-blog\/wp-json\/wp\/v2\/posts\/243\/revisions"}],"predecessor-version":[{"id":1635,"href":"https:\/\/marriott-stats.com\/nigels-blog\/wp-json\/wp\/v2\/posts\/243\/revisions\/1635"}],"wp:attachment":[{"href":"https:\/\/marriott-stats.com\/nigels-blog\/wp-json\/wp\/v2\/media?parent=243"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/marriott-stats.com\/nigels-blog\/wp-json\/wp\/v2\/categories?post=243"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/marriott-stats.com\/nigels-blog\/wp-json\/wp\/v2\/tags?post=243"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}